The goal of this project is to investigate how the appraisal of daily events and coping responses to those events are related to the acquisition of problem drinking patterns in women and men. The relation between stress and drinking has generally been studied in laboratory investigations of the cardiovascular response to specific stressors where alcohol consumption typically occurs prior to the stressor. Although important for ensuring that all subjects experience the same stressor, these studies have considered neither individual differences in stress appraisal nor the relation of appraisal to coping or drinking. other studies have focused on the association between stress and alcohol use, or on the association between coping responses and drinking, whereas fewer investigations have examined the combined effect of stressful events and coping on the use of alcohol. It remains unclear, then, why some individuals may drink in response to some situations but not to others, and why they may feel relieved after drinking sometimes but not at others. The proposed study is designed to evaluate whether certain types of events, appraisals, and coping responses predict level of drinking among subjects, and to determine the temporal relations among appraisal, coping, and drinking within individuals. The proposed investigation is a 56-day daily process study. On a daily basis, 144 subjects will record positive and negative events, rate several properties or characteristics of events that are important in the overall perception of event severity, and indicate what they did in response. Half of the subjects will be women. Subjects will be selected to represent the full range of drinking behavior (i.e., from light drinkers to abuse). The prospective daily design will allow stressful events, appraisals, coping responses, and drinking to be recorded closer to their "real time" occurrences, retrospection biases will be minimized, and the temporal sequencing of events and drinking will be established more confidently. The design makes it possible to untangle the temporal relations among these variables to determine how they may increase the risk for alcohol related problems and dependence. Multivariate repeated measures regressions will be conducted to control the Type I error associated with the number of hypotheses tested. Post-hoc comparisons will then be performed for each type of event (leisure time, family relationships, financial matters, and work). Finally, the temporal relations among the study variables will be investigated using a within- subjects design with an intensive time sampling approach to the recording of daily events.